ChatGPT used for evil: Fake IT worker resumes, misinfo, and cyber-op assist Fake IT workers possibly linked to North Korea, Beijing-backed cyber operatives, and Russian malware slingers are among the baddies using ChatGPT for evil, according to OpenAI's latest threat report. The AI giant said it quashed 10 operations using its chatbot to conduct social engineering and cyber snooping campaigns, …
This post has been deleted by its author
Ruskies are talking about Ukrainian nazis yet they continue to support Nazi symphatisers aroung the world to create internal havoc in other countries. Divide and conquer of sorts.
Of course, Russia has their own neonazi groups but Putin sees them as a useful tool.
Who is not a Nazi in your book ?
Its truly a pathetic human being that uses that word for anything other than its true meaning. Do you have no respect whatsover for those that died during the war. Or is it just a question of me, me ,me ...
Please, I implore you, before deploying that word think about what Nazis truly were and think of all the freedoms that you currently have due to your forefathers who fought and died against them so that you could benefit.
"Who is not a Nazi in your book ?"
The Ukrainian government, Ukrainian population, Ukrainian military. For start.
"Its truly a pathetic human being that uses that word for anything other than its true meaning. Do you have no respect whatsover for those that died during the war. Or is it just a question of me, me ,me ...
What the hell are you talking about? I wrote about 'neonazis' and 'Nazi symphatisers'. You don't seem to understand the difference between words 'neonazi' and 'Nazi'. Educate thyself.
"Please, I implore you, before deploying that word think about what Nazis truly were and think of all the freedoms that you currently have due to your forefathers who fought and died against them so that you could benefit."
Why did you write this important request of yours anonymously? Do you not stand by your words?
"However, the warning posed by fascist regimes needs a strong and meaningful label.
The best legacy is to never forget and to point out unacceptable behaviour immediately."
I wholeheartedly agree, but lets give it it's own name rather than make any confusion with existing regimes.
If we use this as an example then "Antifa" should also be held up to the Neonazi moniker, I don't see the difference between them and the far right, they both appear to be equally dangerous. Strangely enough Antifa seems to be the the more violent between the two.
OpenAI’s report is a magician’s trick: point at North Korean coders and Russian trolls while the real con happens in plain sight.
Corporations use the same tools to run fake job ads, harvest CVs, and micro-target insecurities for profit. Political campaigns generate synthetic outrage at scale, flooding the discourse with weaponised noise. None of this is flagged - because it’s lucrative, legal, and dressed in a suit.
You’re not watching a threat report. You’re watching PR - covering for the fact that AI isn’t disrupting power. It’s entrenching it.
"Look over there!" the article screams - while your data, agency, and attention span vanish.
Why would someone with harm in mind risk exposure by using a traceable online tool with zero confidentiality?
Instead they'd run it locally and airgapped, a decent quantized model is cheaper and more confidential to run offline.
So to me this reads like PR messaging, not actually mitigating anything meaningful.
Right. They can easily post misinfo which many people will read and some will believe. Even if the misinformation post is totally debunked later on, it won't reach everyone.
Death by thousand cuts.
It is much cheaper for Russia to try and bring down other countries to their level instead of investing in their own population. Elite excepted of course.
"Even if the misinformation post is totally debunked later on, it won't reach everyone."
And even if it did reach everyone, many will simply refuse to accept the debunking. It will be brushed off as fake news, deep state propaganda and the like.
Many, many people believe only what they want to believe; little details like, truth, logic and evidence are completely lost on them. See flat earthers, moon landing deniers, believers in various sky faries etc.
Having real, high quality experience is not enough, it needs to be tailored to what the hiring company is asking for.
In my experience most CVs are written purely from the author's perspective, and place great emphasis on what matters to the candidate, rather than thinking "what skills and experience does the job ad ask for, and how do I make it easy for the poor sod reading 200 CVs to pick out mine?"
The people who get the interviews aren't the best candidates, they're the ones who did the least bad job of showing they met the essential criteria.
"showing they met the essential criteria."
I suspect the AI is better trained to read between the lines of the job requirements. Because what is in the job description and ad is never the real story.
Or are companies really find and hire 20 yo with a decade of experience with last year's techno hype?
I've noticed a trend of more scammers targeting businesses using ChatGPT, instead of making a quick buck.
More lengthy enquiries, more convincing documents and harder to convince the business that their 'potential customer' never existed.
By the point they notice something is wrong, they've been phished and many of their colleagues and customers have already been duped or compromised, too. Sales, invoice payments and wages have been diverted without raising suspicion. And some still refuse to believe it was a scam at all.
These are genuine issues because they lower the barrier for people with limited technical skill to engage in harmful activities.
ChatGPT can be used by dunces for evil, just like any other technology.
However lets never lose sight of its massive upside. It enables most users - that's you and I - to work smarter, and be more productive.
The tool itself is neutral—its impact depends on how people use it
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
